Back to directory
MARTI
Free519A Framework for LLM-based Multi-Agent Reinforced Training and Inference
About
MARTI trains and runs multiple AI agents together using reinforcement learning. It's built for teams developing complex multi-agent systems that need to coordinate and improve over time. The framework lets agents learn from each other's actions during both training and deployment.
Key Features
- autogen
- camel
- deepseek-r1
- gemma
- large-language-models
Pricing
FreeOpen source. You supply your own LLM API keys.
Categories
CodingDevOps